self.encoding which we set in the init is only intended
to decode gpg´s stderr which uses a system specific encoding.
if we dont encode the data we pass to python-gnupg ourself, it will fallback and use self.encoding.
This might be of no concern if self.encoding is set to 'utf8' and when we are on Linux
which has a preferred encoding of 'utf8'.
But if we are on Windows the preferred encoding for stderr
is most of the time not 'utf8'. If python-gnupg tries to decode a stderr stream that is for example
encoded with 'cp1252' with our set encoding of 'utf8' this will fail.
The solution is to pre-encode the data before we pass it to python-gnupg, so it does not have to
use self.encoding as a fallback. And set self.encoding='latin1' because latin1 will not yield exceptions
on decoding errors. Also gpg itself will fallback to latin1 as stderr encoding when it cant determine the
preferred encoding of a system.
self.decode_errors is used for something differently, and has no influence on the situation.
Fixes#8644
There are mutliple reasons why this is not a good idea
1. It places work on encryption Plugins, as many encryption attributes inside the
stanza can not be resend again (OMEMO, OTR), so the plugins have to make sure none of
these attr are inside the LMC stanza
2. In general its not obvious for plugin devs that a stanza issued after LMC has to be
treated differently. There should be no negative effects, even when a contributor not knowing
about LMC at all.
This commit saves only the stanza id, and adds the replace tag on the new message.
This results also in less code.
We have to save the value per account and contact, because otherwise
it could lead to problems if a jid is added to more than one account.
As key this uses a string in the form of 'Account-BareJid'.
The config option group 'contacts' is not used because there are values
like 'speller_language' that make more sense to set on a Jid basis, rather than a
Account-Jid basis.
Treat an empty config value or no config value as 'disabled', instead
of writing a config value of 'disabled'. This has the benefit that we
dont have to write config values for contacts were we dont use encryption.
- Add extension points when receiving/sending a message
- Add extension point for setting the lock image
- Add extension point after clicking the send button
- Add extension point when typing
- Add a new menu button to choose the desired encryption
- Extend the PluginManager to hold a list of encryption plugins
Use the account hostname instead of the target hostname of the
_xmpp-client._tcp SRV resource record for STUN server discovery.
Otherwise, discovery fails if both hostnames are different.
If the user had an empty `file_transfer_proxies` config setting
no proxies were used. As we discover the default server proxy ourself the
user should not have to write it to his config setting to make use of it.
- We want to send requests also to offline contacts.
- The XEP is wide spread chances are high that a contact supports it.
- It gets really complicated, when we want to guess if a offline contact supports receipts
- 'received'-Carbons from ourself, because we either received
the message directly or got a 'sent'-Carbon about it
- Messages from our own full jid, this can happen when we send
a Message to ourself and no other resource is online.
- Push Invitation and Decline event much earlier
- Refactor Logic to decide if mediated or not
Ejabbered 17.03 does add a 'jabber❌conference' namespace into mediated invites which has broken our logic
- Dont add a groupchat contact on invitation, always add the inviter to the roster
- Move handling of Data IQs into IBBIqHandler
- Call SendHandler with file_props
- Save last sent id in file_props.syn_id
- Remove some useless checks, now that we call SendHandler with file_props
- Send item-not-found error on invalid session id
Emmanuel Gil Peyrot
Yann Leboulanger
Philipp Hörist
André Apitzsch
Marc Schink
Yann Leboulanger